Font Size: a A A

Research On Electric Fire Risk Assessment Based On Data Mining Technology

Posted on:2017-04-25Degree:MasterType:Thesis
Country:ChinaCandidate:Z C ZhangFull Text:PDF
GTID:2131330482988400Subject:Safety science and engineering
Abstract/Summary:PDF Full Text Request
Currently, as one of the main energies.electric energy facilitates human production and human life, but also quietly influences all aspects of production and life as a potential risk source. Electrical fire is featured by high frequency, big economic loss, and strong elusiveness as the first reason of fire. Along with the fast development of data acquisition and data storage technologies, huge amount of electrical fire data is accumulated as the data base for the analysis on rule of electrical fire. Meanwhile, data mining technology theories and methods are effective to explore and analyze electrical fire accident.Through the statistical analysis on electrical fire accident data from 2004 to 2014, data mining technologies including decision tree analysis, word frequency analysis, and social network analysis are used to analyze the deep reasons for great electrical fire accident and screen the key indicators for electrical fire risk assessment. Based on electric safety management survey investigation on production and operation companies in Beijing, with reference to 4M and PDCA in quality management system and laws, regulations and standards related to electrical fire, electrical fire assessment indicator system is constructed. Through the comparative analysis, AHP-entropy method is used to determine the weights of electrical fire risk assessment indicators and construct the electrical fire risk calculation method upon man risk, fixed risk, environmental risk, and safety management risk. Electrical fire risk is classified into 4 levels:high risk, relatively high risk, intermediate risk, low risk.Electrical fire risk assessment model is applied to the enterprise management.It is found in the study that the enterprise’s electrical fire risk value is 34.46, which is intermediate risk. By means of MATLAB software programming, the three-level indicators are shown in the electrical fire risk radar map. Man behavior,equipped facility and fire-fighting equipment are found to be dangerous clearly. Pertinent counter measures and suggestions are proposed upon the result of electrical fire risk assessment result of the enterprise. Thus,focus on security issues.prevent electrical fires,reduce casualties and economic losses,and is of important guidance and realistic value for enterprise.
Keywords/Search Tags:Electrical fire, Risk assessment, Data mining, Index system, Analysis hierarchy entropy
PDF Full Text Request
Related items